Small Sample Tests for Serial Correlation in Models Containing Lagged Dependent Variables

1975 ◽  
Vol 57 (3) ◽  
pp. 383 ◽  
Author(s):  
James L. Kenkel
Keyword(s):  
Serial Correlation ◽  
Small Sample ◽  
Dependent Variables

Comparing the Performance of Multivariate Multilevel Modeling to Traditional Analyses With Complete and Incomplete Data

Methodology ◽  
2015 ◽  
Vol 11 (3) ◽  
pp. 100-109 ◽  
Author(s):  
Ryoungsun Park ◽  
Keenan A. Pituch ◽  
Jiseon Kim ◽  
Hyewon Chung ◽  
Barbara G. Dodd
Keyword(s):  
Missing Data ◽  
Multilevel Modeling ◽  
Error Rate ◽  
Type I Error ◽  
Small Sample Size ◽  
Univariate Analysis ◽  
Small Sample ◽  
Type I ◽  
Type I Error Rate ◽  
Dependent Variables

Abstract. A multivariate multilevel model (MVMM) extends standard multilevel modeling by including multiple dependent variables and thus could be used in place of traditional multivariate analyses. For a two-group study with two correlated dependent variables, a simulation study was conducted to compare the performance of MVMM to traditional MANOVA and a series of univariate analyses. The results showed that MVMM provides for virtually always greater power than other analyses, even for conditions that have been previously shown to favor univariate analysis. Further, this power advantage can be substantial even when no missing data are present and is often much greater when data are missing. While the Type I error rate for the overall multivariate null hypothesis can be somewhat elevated with MVMM, especially with small sample size and a large proportion of missing data, the Type I error rate for the test of a specific dependent variable is accurate.

A new approach for estimating a nonlinear growth component in multilevel modeling

2011 ◽  
Vol 35 (4) ◽  
pp. 370-379 ◽  
Author(s):  
Asko Tolvanen ◽  
Noona Kiuru ◽  
Esko Leskinen ◽  
Kai Hakkarainen ◽  
Mikko Inkinen ◽  
...  
Keyword(s):  
Multilevel Modeling ◽  
Small Sample ◽  
Monte Carlo Simulation Study ◽  
New Approach ◽  
Nonlinear Growth ◽  
Fixed And Random Effects ◽  
Dependent Variables ◽  
Random Coefficient Regression ◽  
Day Of The Week ◽  
Small Sample Sizes

This study presents a new approach to estimation of a nonlinear growth curve component with fixed and random effects in multilevel modeling. This approach can be used to estimate change in longitudinal data, such as day-of-the-week fluctuation. The motivation of the new approach is to avoid spurious estimates in a random coefficient regression model due to the synchronized periodical effect (e.g., day-of-the-week fluctuation) appearing both in independent and dependent variables. First, the new approach is introduced. Second, a Monte Carlo simulation study is carried out to examine the functioning of the proposed new approach in the case of small sample sizes. Third, the use of the approach is illustrated by using an empirical example.

Sign in / Sign up

Export Citation Format

Share Document